[android-common:android16-6.12-p3s 2/2] arch/arm64/kvm/mmu.c:378: warning: Function parameter or struct member 'addr' not described in '___unmap_stage2_range'

[kernel test robot <lkp@intel.com>]

Hi Quentin,

FYI, the error/warning still remains.

tree:   https://android.googlesource.com/kernel/common android16-6.12-p3s
head:   606d03f6372c0cd085fc0d154cbfc17c298058f2
commit: fcc111df8789eb3964c040712266abb15c823476 [2/2] ANDROID: KVM: arm64: Unshare pages from __unmap_stage2_range()
config: arm64-allmodconfig (https://download.01.org/0day-ci/archive/20261029/202610292257.ELEb1wQN-lkp@intel.com/config)
compiler: clang version 24.0.0git (https://github.com/llvm/llvm-project c2ea4440d4b64c148de4fdfca1fc2c437a0701e4)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20261029/202610292257.ELEb1wQN-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202610292257.ELEb1wQN-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> arch/arm64/kvm/mmu.c:378: warning: Function parameter or struct member 'addr' not described in '___unmap_stage2_range'
>> arch/arm64/kvm/mmu.c:378: warning: expecting prototype for __unmap_stage2_range(). Prototype was for ___unmap_stage2_range() instead

Kconfig warnings: (for reference only)
   WARNING: unmet direct dependencies detected for INTEL_MEI_ME
   Depends on [n]: X86 && PCI [=y]
   Selected by [m]:
   - INTEL_MEI_HDCP [=m] && (DRM_I915 [=n] || DRM_XE [=m])
   - INTEL_MEI_PXP [=m] && (DRM_I915 [=n] || DRM_XE [=m])


vim +378 arch/arm64/kvm/mmu.c

fcc111df8789eb arch/arm64/kvm/mmu.c Quentin Perret   2022-07-06  339  
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  340  /*
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  341   * Unmapping vs dcache management:
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  342   *
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  343   * If a guest maps certain memory pages as uncached, all writes will
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  344   * bypass the data cache and go directly to RAM.  However, the CPUs
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  345   * can still speculate reads (not writes) and fill cache lines with
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  346   * data.
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  347   *
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  348   * Those cache lines will be *clean* cache lines though, so a
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  349   * clean+invalidate operation is equivalent to an invalidate
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  350   * operation, because no cache lines are marked dirty.
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  351   *
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  352   * Those clean cache lines could be filled prior to an uncached write
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  353   * by the guest, and the cache coherent IO subsystem would therefore
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  354   * end up writing old data to disk.
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  355   *
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  356   * This is why right after unmapping a page/section and invalidating
52bae936f0e7be arch/arm64/kvm/mmu.c Will Deacon      2020-09-11  357   * the corresponding TLBs, we flush to make sure the IO subsystem will
52bae936f0e7be arch/arm64/kvm/mmu.c Will Deacon      2020-09-11  358   * never hit in the cache.
e48d53a91f6e90 virt/kvm/arm/mmu.c   Marc Zyngier     2018-04-06  359   *
e48d53a91f6e90 virt/kvm/arm/mmu.c   Marc Zyngier     2018-04-06  360   * This is all avoided on systems that have ARM64_HAS_STAGE2_FWB, as
e48d53a91f6e90 virt/kvm/arm/mmu.c   Marc Zyngier     2018-04-06  361   * we then fully enforce cacheability of RAM, no matter what the guest
e48d53a91f6e90 virt/kvm/arm/mmu.c   Marc Zyngier     2018-04-06  362   * does.
363ef89f8e9bce arch/arm/kvm/mmu.c   Marc Zyngier     2014-12-19  363   */
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  364  /**
ffd9eaffa34d79 arch/arm64/kvm/mmu.c Randy Dunlap     2024-01-17  365   * __unmap_stage2_range -- Clear stage2 page table entries to unmap a range
c9c0279cc02b4e arch/arm64/kvm/mmu.c Xiaofei Tan      2020-09-17  366   * @mmu:   The KVM stage-2 MMU pointer
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  367   * @start: The intermediate physical base address of the range to unmap
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  368   * @size:  The size of the area to unmap
c9c0279cc02b4e arch/arm64/kvm/mmu.c Xiaofei Tan      2020-09-17  369   * @may_block: Whether or not we are permitted to block
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  370   *
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  371   * Clear a range of stage-2 mappings, lowering the various ref-counts.  Must
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  372   * be called while holding mmu_lock (unless for freeing the stage2 pgd before
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  373   * destroying the VM), otherwise another faulting VCPU may come in and mess
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  374   * with things behind our backs.
7a1c831ee8553b arch/arm/kvm/mmu.c   Suzuki K Poulose 2016-03-23  375   */
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11  376  
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11  377  static int ___unmap_stage2_range(struct kvm_s2_mmu *mmu, u64 addr, u64 size)
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11 @378  {
fcc111df8789eb arch/arm64/kvm/mmu.c Quentin Perret   2022-07-06  379  	if (!is_protected_kvm_enabled())
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11  380  		return kvm_pgtable_stage2_unmap(mmu->pgt, addr, size);
fcc111df8789eb arch/arm64/kvm/mmu.c Quentin Perret   2022-07-06  381  
fcc111df8789eb arch/arm64/kvm/mmu.c Quentin Perret   2022-07-06  382  	return pkvm_unmap_range(kvm_s2_mmu_to_kvm(mmu), addr, addr + size);
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11  383  }
80a72f6ae4aa66 arch/arm64/kvm/mmu.c Quentin Perret   2022-05-11  384  

:::::: The code at line 378 was first introduced by commit
:::::: 80a72f6ae4aa66b8fb00a1c8a0d61f054d618003 ANDROID: KVM: amr64: Pass kvm_s2_mmu struct from stage2_apply_range()

:::::: TO: Quentin Perret <qperret@google.com>
:::::: CC: Keir Fraser <keirf@google.com>

--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki